CA1258565A - Process for purifying industrial gases and industrial flue gases - Google Patents
Process for purifying industrial gases and industrial flue gasesInfo
- Publication number
- CA1258565A CA1258565A CA000519215A CA519215A CA1258565A CA 1258565 A CA1258565 A CA 1258565A CA 000519215 A CA000519215 A CA 000519215A CA 519215 A CA519215 A CA 519215A CA 1258565 A CA1258565 A CA 1258565A
- Authority
- CA
- Canada
- Prior art keywords
- process according
- wash water
- formaldehyde
- oxidizing agent
- cyanide
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/14—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols by absorption
- B01D53/1412—Controlling the absorption process
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/14—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols by absorption
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/14—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols by absorption
- B01D53/1493—Selection of liquid materials for use as absorbents
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/30—Controlling by gas-analysis apparatus
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/34—Chemical or biological purification of waste gases
- B01D53/46—Removing components of defined structure
- B01D53/54—Nitrogen compounds
Abstract
ABSTRACT OF THE DISCLOSURE
In the purification of cyanide-containing gas wash water with formaldehyde the accurate dosage of formaldehyde constitutes a problem particularly in the case of intensely varying cyanide contents. By continuous potentiometric measurement in a secondary flow which is adjusted to a pH
value of 7 to 10 while the pH value of the main flow remains unchanged the formaldehyde dosage can be reliably adapted to the cyanide contents at any given time.
In the purification of cyanide-containing gas wash water with formaldehyde the accurate dosage of formaldehyde constitutes a problem particularly in the case of intensely varying cyanide contents. By continuous potentiometric measurement in a secondary flow which is adjusted to a pH
value of 7 to 10 while the pH value of the main flow remains unchanged the formaldehyde dosage can be reliably adapted to the cyanide contents at any given time.
Description
1~58S~iS
The present invention relates to a process for puri~
fying industrial gases such as industrial flue gases which contain hydrogen cyanide but frequently also ammonia and in which oxides of carbon hydrocarbons of various structures, hydrogen sulphlde, sulphur oxides and dusts containing heavy metals are present.
These gases are obtained, for example, in blast fur-nace processes, acrylonitrile syntheses, garbage pyrolyses and other chemical processes and are purified by washing with aqueous formaldehyde. The present invention relates specifi-cally to the electrometrically controlled dosage of the formaldehyde.
The purification of industrial gases or industrial flue gases is required for various reasons. It serves for separating dusts from utilizable gases, like those obtained, in blast furnace processes and garbage pyrolyses on the one hand. Flue gases formed, for example, in the acrylonitrile syntheses and in other chemical syntheses on the other must be freed from deleterious substances prior to their release into the atmosphere. Heretofore blast-furnace gases were frequently freed from dusts in the continuous-flow process.
After its use the wash water was passed over a circular thickener simultaneously serving as a detoxication basin and the clear decantate was passed into the sewer system or into a main canal. However, since very large quantities of water are required for this purpose, the circulation process is pre-ferred. After washing the gas the wash water is mixed with flocculating agents, passed over a circular thickener and the clear decantate is again passed to the washing cycle. Fresh water is added only to co~pensate for losses due to evapora-tion and to reduce the hardness of the water. Only a fraction of the wash water thus gets to the detoxication, but it has ~
1;~58565 substantially higher content of deleterious substances than the wash water from the non-circulating water cooling. The pH
of the wash waters in the cycle is adjusted in the rarest cases so that the necessarily obtained pH value is in the neu-tral range. This has the result that a major portion of the cyanide is discharged from the evaporation coolers in the form of hydrogen cyanide.
DE-OS 24 60 927 describes a two-stage process for treating blast-furnace-gas wash water. This process is based on the fact that the cyanide content is measured in front of the sedimentation installations and that 20 to 70% of the amount of formaldehyde required for the stoichiometric reac-tion to glycolic acid nitrile are added. The pH of the blast-furnace-gas wash water must be adjusted in this case to value of 8 to 10 in its entirety.
A very thorough mixing in the region of addition of formaldehyde is important for the process in order to avoid local over concentration of formaldehyde which can result in the reaction with other substances, as for example, ammonium ions. However, excesses of formaldehyde, i.e., also local excesses, must be avoided in any case. A specialty of the process is seen in this.
In a second stage the remalning portion of the stoi-chiometric amount of formaldehyde, which is required for the reaction with hydrogen cyanide, is then added after the gra-vity separator and after removing the glycol nitrile polymers formed in the first stage. However, for this type of process exact data for the control of the formaldehyde addition would be required, for example, in order to avoid the allegedly inadmissible local excesses of formaldehyde in the presence of reactive ions, for example, ammonium ions, but these data are totally lacking. However, in view of the intensely varying lZ5~356S
cyanide content in the wash water of blast furnace gases, for example, between 0.1 and 202 mg of CN per litre of wash water depending on the production process for iron, these data are required since dosing according empirical values can then no longer be carried out, quite apart from the fact that appre-ciable variations in the cyanide content can also occur during the production itself. A process for the detoxication of intensely cyanide-containing effluents with an alkaline formaldehyde solution whose pH value is at least 8 and which is preferably applied in excess over the stoichiometrically required amount is described in DE-OS 21 19 119. This process is expensive to carry out either by heating the effluent for several hours to the boiling temperature or by allowing it to stand for several days at room temperature. An adequate detoxication is not attained in any of these cases. Thus, for example, after boiling the effluent for two hours the residual cyanide content is 0.5 mg of CN per litre and after allowing the effluent to stand for 50 hours at room temperature it is 8 mg of CN per litre.
A process for the decontamination primarily of effluents of acrylonitrile plants which are subsequently passed to a biological clarifying plant comprises the use of a formaldehyde solution having a pH value of 3 or lower (DE-PS
22 02 660), that is to say, in molar excess. Preferably 1.5 to 4 moles of formaldehyde are used per mole of cyanide ion.
However, this continuous process can be checked only by a wet analytical determination of the cyanide content in the effluent, since according to data in said patent the adjustment of the equilibrium is so fast that it can no longer be measured with a silver iodide electrode. However, it is not evident from the examples how the intensely varying amount of effluents - amounts of between 20 and 40 cu m per hour hav-1.i~58565 ing cyanide values 20 to 300 p.p.m. are mentioned - can be detoxicated without electrometric control with no problems.
In any case it is evident from the examples that even when applying four times the amount of formaldehyde, relative to the cyanide concentration, the limiting value of <0.1 mg of CN per litre required today is not attained in any case.
~ccording to this process no glycol nitrile is to be formed but an unknown reaction is assumed to proceed, probably while forming pyrimidones. Therefore, it is the aim of the present invention to provide a process in which hydrogen cyanide in gases and/or gas wash waters is quantitatively converted into glycolic acid nitrile by adding technical commercial formalde-hyde under electrometric control without having to apply a large excess of formaldehyde, relative to cyanide, whereupon a detoxication of the glycol nitrile follows.
Control of the formaldehyde dosage by means of a direct measurement of the redox potential in the effluent is excluded since in the reaction of formaldehyde with cyanide in the pH range from 1 to 12 the influence of the pH differences on the redox potential is too great.
It has now been found that the purification of industrial gases such as industrial flue gases, which contain hydrogen cyanide and preferably ammonia, and is carried out by cycle washing with water and formaldehyde and a simultaneous oxidative treatment of the wash water removed from the circu-lation followed in both cases by a biological after purifica-tion, can be potentiometrically observed and controlled in the presence of at least 1 ppb of silver ions when the formalde-hyde is dosed into the gas washing cycle ahead of the gas washer and the amount added is regulated by means of a contin-uous measurement of the redox potential with a pair of elec-trodes, comprising a noble metal electrode and a reference ~ '~ S~ 56 ~
electrode in a continuous secondary flow (i.e., measuring flow) branched off after the washes, the pH value of said secondary flow being adjusted by adding liquor or acid to a constant value of between 7 and 10, preferably 8 to 8.5. The gases to be purified can additionally contain oxides of car-bon, hydrocarbons of various structures, hydrogen sulphide oxides of sulphur, and dusts containing heavy metals. The measuring flow is preferably removed between gas washer and sedimentation installation. This small measuring flow, approximately 100 litres per hour, is branched off and ad~usted to the constant pH value by automatically regulated addition of liquor or acid. The solution of a stable silver compound is simultaneously dosed in, when required, together with the ac~d. A silver concentration of at least 1 ppb of silver ions is maintained in the measuring flow. A preferred upper limiting value is 10 ppb of silver ions. However, 1 ppb of silver ions usually is sufficient for carrying out the measurement. Higher silver concentrations, for example, 100 mg per cu m, do not interfere with the process but they can have a detrimental effect in a biological clarification plant or in a natural watercourse. According to the present invention the main flow is not changed with regard to its p~
value.
Corresponding to the redox potential measured in the measuring flow there is controlled, via an electronic regula-tor with P, PI or PID characteristics, a dosing pump which doses the corresponding amount of a commercial formaldehyde solution into the ring conduit of the main flow ahead of the gas washer. Diluted or gaseous formaldehyde can also be used instead of the commercial formaldehyde quantities. When using, for example, a pai~ of gold thallamide electrodes for this purpose, the redox potential in the measuring flow 1;~5856S
reaches a value ranging from approximately +400 mv + 50 mv to ~100 mv + 50 mV in a pH value range from 7 to 10, preferably 8 to 8.5 as soon as the reaction of the cyanide with the formaldehyde is completed.
A preferred rated redox value lies between +650 mv +
50 mV and +850 mV + 50 mv, particularly preferred at +700 mV +
50 mv at a pH value of 8 to 8.5. When using a silver elec-trode instead of a gold electrode the rated redox values are lower by approximately 100 mV on reacting the end product of the reaction between cyanogen and formaldehyde. The optimal ad~ustment of the rated value must be determined by a preced-ing small-scale test this also applies correspondingly to other combinations of electrode pairs. By adding the formaldehyde solution ahead of the gas washer and by measuring the redox potential in the measuring flow after the gas washes the hydrogen cyanide contained in the gas to be washed is immediately converted into glycolic acid nitrile on the one hand but a reaction of the hydrogen cyanide with metals while forming heavy metal cyanides which are difficult to detoxicate is avoided on the other. Furthermore, at a high content of ammonia in the wash water hexamethylene tetramine can be formed with formaldehyde. The hexamethylene tetramine again releases formaldehyde to the extent that it is required for the reaction with the hydrogen cyanide while forming glycolic acid nitrile.
For this reaction a reaction time of maximally 2 minutes is required. This can always be maintained by means of the arrangement of the measuring and dosing points in the process according to the present invention. Furthermore, in this manner it is assured that no free formaldehyde is present in the washing cycle since only the amount stoichiometric to the cyanide content is dosed in by the redox-controlled dosage 1;~58565 amount on the one hand and in the presence of high NH3 con-tents an unintended slight excess is converted into biologi-cally readily degradeable hexamethylene tetramine on the other. For the decomposition of the glycol nitrile formed an oxidative treatment of the washing cycle or of the wash water J removed from the circulation is carried out.
Primarily hydrogen peroxide is suitable as oxidizing agent. Hydrogen-peroxide solutions in concentrations of 10 to 70% by weight are preferably used. A further important effect attained by the process according to the present invention is the reduction of the discharge of hydrogen cyanide from the evaporation coolers. The cooling cycles operated in foundries must be fitted with evaporation coolers to maintain an ade-quate cooling water temperature. In a laboratory test adjusted to plant conditions it could be determined that a flue gas with 15 p.p.m. of hydrogen cyanide results from a blast-furnace-gas wash water loaded with 10 mg of CN per litre while the flue gas from a blast-furnace-gas wash water treated with formaldehyde and having the same initial cyanide concentration contained only 2 p.p.m. of hydrogen cyanide.
secause of the reconcentration of the wash water and the deposits in the gas washers associated therewith it is very favourable to remove contlnuously a small amount of wash water from the circulation. Instead of the washing cycle this discharged water is ad~usted with lime or alkali liquor to pH
value of 8.5 to 12.5, preferably 10.5 and controlled poten-tiometrically, for example, adding H202, until the redox potential of +700 mV + 50 mV, as measured with a pair of gold-thallium amalgam/thallium chloride electrodes is attained.
The change in potential thus occurring can be used for signalizing the end of the detoxication reaction or in the case of continuous dosing of oxidizing agent for switching off l;~S~35~5 the dosing operation. As mentioned hereinbefore, a biological treatment, for example, in a correspondingly adapted clarifl-cation plant, can follow.
The process according to the present invention is applicable primarily to inorganic cyanides obtained in wash waters of smelting operations, garbage pyrolyses and plants processing or producing hydrocyanic acid or cyanide. Since the measuring flow is branched off at the point of the strongest cyanide ions concentration of the wash water the addition of the formaldehyde into the washing cycle prior to entering the gas washer is always so regulated that this addi-tion also is adequate for varying concentrations of cyanide ions. This is a decisive advantage of the process according to the present invention. Furthermore, the gas washers can be operated in pH ranges in which the carbon dioxide of the air is not absorbed.
Heretofore, the operations had to be carried out in the pH range >13 to assure a complete separation of the hydro-gen cyanide and of the carbon dioxide. This is required since the carbonic acid is a stronger acid than the hydrocyanic acid and in the lower pH range carbon dioxide is preferably absorbed by the wash water. Thus, according to the process of the present invention a distinctly lower load of the wash water with neutral salts is caused than was the case hereto-fore. The pH range of the wash water does not have to be changed for reasons of an undesired absorption of gases, as for example, the absorption of carbon dioxide, nor for reasons of effectively carrying out the process according to the present invention. The pH range can vary within wide limits but it will usually be below pH = 7. However, in individual cases, as in the treatmen~t of flue gases and in the production 1;~5856S
of ferromanganese, values of between 10 and 11 are also possible.
All of these effluents can be used for the process according to the present invention with the pH value unchanged. The measuring flow alone is adjusted to a pH value of preferably 8 to 8.5. It should also be emphasized that even large amounts of ammonium ions sulphide ions, as they are encountered in effluents from the iron and steel industry and garbage pyrolyses, do not interfere with the measurement in the measuring flow (for the determination of the influence of ammonia on the required amount of formaldehyde see the Exam-ples 3 and 4). The following two principles of carrying out the process serve for further illustrating the process accord-ing to the present invention: these two procedures (see Figure 1 and 2) can be carried out individually or in combina-tion - adapted to any given case. Thus, for example, the method according to Figure 1 can be used instead of the wash-ing cycle in Figure 2, i.e., instead of the system '~gas washer - measuring flow - actual washing cycle". In this latter case dosing of oxidizing agent into the washing cycle is then dis-pensed with. In Figure 1 the washing of gas with formaldehyde and simultaneous oxidation with hydrogen peroxide in the cir-culation system is shown. The crude gas is fed via llne 1 to the gas washer 2 from below and the pure gas is removed via line 3. The wash water is fed via line 4 to the gas washer 2 from above in a counterflow and drawn off via line 16.
The suspended substances contained in the wash water are separated in the sedimentation tank 5, for example, a cir-cular reamer, and the wash water is then passed via the line 17 to the cooler 6, for example, an evaporation cooler, from where it gets back into the gas washer 2 via line 4.
125~356~
The washing cycle comprises the lines 4,16,17 as well as the gas washer 2, the sedimenta-tion tank 5 and the cooler 6. After the gas washer 2 a small measuring flow 7 is branched off from the washing cycle and controlled via a regu-lator 8. The pH value of the measuring flow is adjusted to, e.g., 8 to 8.5 by adding acid via line 9 or liquor via line 10. Corresponding to the redox potential also measured with the regulator 8 and to the deviation thus determined formalde-hyde is dosed via line 11 into the washing cycle prior to entering the washer 2. The silver compound, for example, nitrate, enters the measuring flow 7 via line 9a.
The redox potential of the washing cycle, whose pH
value is unchanged, is measured with the regulator 12 and cor-responding to the deviation determined hydrogen peroxide is dosed via line 13 after the circular reamer 5. The residence time in the cooler and in the entire washing cycle is utilized for the perhydrolysis of the glyconitrile formed in the gas washer 2. The water washed from the cooler 6, whose pH must be changed when required and is drawn off via line 14, is replaced by adding the corresponding amount of fresh water via line 15 after the cooler 6.
In Figure 2 the washing of the gas with formaldehyde and separate treatment of gas wash water with hydrogen per-oxide is shown. The crude gas is fed via line 1 to the gas washer 2 from below and the pure gas is drawn off via line 3.
The wash water is passed in a counterflow via line 4 to the gas washer 2 from above and drawn off via line 4a. The sus-pended substances contained in the wash water are removed in the sedimentation tank 5 and then passed to the detoxication plant.
After the gas washer 2 a small measuring flow 7 is taken out of the washing cycle via line 16 and controlled via ~58565 a regulator 8 by adding liquor via line 9 or acid via line 10 and adjusted to a pH value of 8 to 8.5 while adding a silver compound via line 9.
Corresponding to both the redox potential also measured with the regulator 8 and the deviation thus deter-mined formaldehyde is dosed via line 11 into the washing cycle, i.e., into line 4, prior to entering the washer 2. At the same time as much fresh water is added via pipe 12 ahead of the gas washer 2 as is washed away via line 5. The ratio of water washed away to the amount of water in the washing cycle is determined by both the degree of pollution and the degree of hardness. This can be d~termined by a preliminary test.
The water discharged from the sedimentation device 5 is first passed to the first reaction tank 13, which is pro-vided with a stirrer 14, a pH-measuring and control device 15 as well as with a liquor dosing device 16 and an H2O2 dosing device 17. When required, the pH of the effluent to be treated is first adjusted to a pH value >10.5. In the second reaction tank 18 provided with a stirrer 19 and a redox-measuring and control device 20 the corresponding amount of H2O2 is dosed via 17 into the first reaction tank 13 according to both the measured redox potential and the deviation thus determined. The size of the reaction tanks is such, that an adequate residence time corresponding to the effluent flow is attained in order to assure a complete perhydrolysis of the glyconitrile. The reaction is potentiometrically monitored via the regulator 20.
The oxidatively treated water is then passed into the third reaction tank 21, which is provided with a strirrer 22 a pH-measuring and control device 23 and an acid dosing l~S~5~jS
device 24. The effluent is then adjusted to a pH value pre-scribed for the discharge.
The present invention will be further illustrated by the following Examples:
The effect which can be attained by a potentiometri-cally controlled addition of formaldehyde is described in Example 1.
A washing cycle having a volume of 5 cu m and a cir-culation of 60 cu m per hour had the following composition:
directly argentometrically determinable cyanide83 mg of CN /litre total cyanide DIN 38 405 D13.1135 mg of CN /litre free N~3 3800 mg of NH3/litre total NH3 3800 mg of NH3/litre iron, total 46.5 mg of Fe/litre
The present invention relates to a process for puri~
fying industrial gases such as industrial flue gases which contain hydrogen cyanide but frequently also ammonia and in which oxides of carbon hydrocarbons of various structures, hydrogen sulphlde, sulphur oxides and dusts containing heavy metals are present.
These gases are obtained, for example, in blast fur-nace processes, acrylonitrile syntheses, garbage pyrolyses and other chemical processes and are purified by washing with aqueous formaldehyde. The present invention relates specifi-cally to the electrometrically controlled dosage of the formaldehyde.
The purification of industrial gases or industrial flue gases is required for various reasons. It serves for separating dusts from utilizable gases, like those obtained, in blast furnace processes and garbage pyrolyses on the one hand. Flue gases formed, for example, in the acrylonitrile syntheses and in other chemical syntheses on the other must be freed from deleterious substances prior to their release into the atmosphere. Heretofore blast-furnace gases were frequently freed from dusts in the continuous-flow process.
After its use the wash water was passed over a circular thickener simultaneously serving as a detoxication basin and the clear decantate was passed into the sewer system or into a main canal. However, since very large quantities of water are required for this purpose, the circulation process is pre-ferred. After washing the gas the wash water is mixed with flocculating agents, passed over a circular thickener and the clear decantate is again passed to the washing cycle. Fresh water is added only to co~pensate for losses due to evapora-tion and to reduce the hardness of the water. Only a fraction of the wash water thus gets to the detoxication, but it has ~
1;~58565 substantially higher content of deleterious substances than the wash water from the non-circulating water cooling. The pH
of the wash waters in the cycle is adjusted in the rarest cases so that the necessarily obtained pH value is in the neu-tral range. This has the result that a major portion of the cyanide is discharged from the evaporation coolers in the form of hydrogen cyanide.
DE-OS 24 60 927 describes a two-stage process for treating blast-furnace-gas wash water. This process is based on the fact that the cyanide content is measured in front of the sedimentation installations and that 20 to 70% of the amount of formaldehyde required for the stoichiometric reac-tion to glycolic acid nitrile are added. The pH of the blast-furnace-gas wash water must be adjusted in this case to value of 8 to 10 in its entirety.
A very thorough mixing in the region of addition of formaldehyde is important for the process in order to avoid local over concentration of formaldehyde which can result in the reaction with other substances, as for example, ammonium ions. However, excesses of formaldehyde, i.e., also local excesses, must be avoided in any case. A specialty of the process is seen in this.
In a second stage the remalning portion of the stoi-chiometric amount of formaldehyde, which is required for the reaction with hydrogen cyanide, is then added after the gra-vity separator and after removing the glycol nitrile polymers formed in the first stage. However, for this type of process exact data for the control of the formaldehyde addition would be required, for example, in order to avoid the allegedly inadmissible local excesses of formaldehyde in the presence of reactive ions, for example, ammonium ions, but these data are totally lacking. However, in view of the intensely varying lZ5~356S
cyanide content in the wash water of blast furnace gases, for example, between 0.1 and 202 mg of CN per litre of wash water depending on the production process for iron, these data are required since dosing according empirical values can then no longer be carried out, quite apart from the fact that appre-ciable variations in the cyanide content can also occur during the production itself. A process for the detoxication of intensely cyanide-containing effluents with an alkaline formaldehyde solution whose pH value is at least 8 and which is preferably applied in excess over the stoichiometrically required amount is described in DE-OS 21 19 119. This process is expensive to carry out either by heating the effluent for several hours to the boiling temperature or by allowing it to stand for several days at room temperature. An adequate detoxication is not attained in any of these cases. Thus, for example, after boiling the effluent for two hours the residual cyanide content is 0.5 mg of CN per litre and after allowing the effluent to stand for 50 hours at room temperature it is 8 mg of CN per litre.
A process for the decontamination primarily of effluents of acrylonitrile plants which are subsequently passed to a biological clarifying plant comprises the use of a formaldehyde solution having a pH value of 3 or lower (DE-PS
22 02 660), that is to say, in molar excess. Preferably 1.5 to 4 moles of formaldehyde are used per mole of cyanide ion.
However, this continuous process can be checked only by a wet analytical determination of the cyanide content in the effluent, since according to data in said patent the adjustment of the equilibrium is so fast that it can no longer be measured with a silver iodide electrode. However, it is not evident from the examples how the intensely varying amount of effluents - amounts of between 20 and 40 cu m per hour hav-1.i~58565 ing cyanide values 20 to 300 p.p.m. are mentioned - can be detoxicated without electrometric control with no problems.
In any case it is evident from the examples that even when applying four times the amount of formaldehyde, relative to the cyanide concentration, the limiting value of <0.1 mg of CN per litre required today is not attained in any case.
~ccording to this process no glycol nitrile is to be formed but an unknown reaction is assumed to proceed, probably while forming pyrimidones. Therefore, it is the aim of the present invention to provide a process in which hydrogen cyanide in gases and/or gas wash waters is quantitatively converted into glycolic acid nitrile by adding technical commercial formalde-hyde under electrometric control without having to apply a large excess of formaldehyde, relative to cyanide, whereupon a detoxication of the glycol nitrile follows.
Control of the formaldehyde dosage by means of a direct measurement of the redox potential in the effluent is excluded since in the reaction of formaldehyde with cyanide in the pH range from 1 to 12 the influence of the pH differences on the redox potential is too great.
It has now been found that the purification of industrial gases such as industrial flue gases, which contain hydrogen cyanide and preferably ammonia, and is carried out by cycle washing with water and formaldehyde and a simultaneous oxidative treatment of the wash water removed from the circu-lation followed in both cases by a biological after purifica-tion, can be potentiometrically observed and controlled in the presence of at least 1 ppb of silver ions when the formalde-hyde is dosed into the gas washing cycle ahead of the gas washer and the amount added is regulated by means of a contin-uous measurement of the redox potential with a pair of elec-trodes, comprising a noble metal electrode and a reference ~ '~ S~ 56 ~
electrode in a continuous secondary flow (i.e., measuring flow) branched off after the washes, the pH value of said secondary flow being adjusted by adding liquor or acid to a constant value of between 7 and 10, preferably 8 to 8.5. The gases to be purified can additionally contain oxides of car-bon, hydrocarbons of various structures, hydrogen sulphide oxides of sulphur, and dusts containing heavy metals. The measuring flow is preferably removed between gas washer and sedimentation installation. This small measuring flow, approximately 100 litres per hour, is branched off and ad~usted to the constant pH value by automatically regulated addition of liquor or acid. The solution of a stable silver compound is simultaneously dosed in, when required, together with the ac~d. A silver concentration of at least 1 ppb of silver ions is maintained in the measuring flow. A preferred upper limiting value is 10 ppb of silver ions. However, 1 ppb of silver ions usually is sufficient for carrying out the measurement. Higher silver concentrations, for example, 100 mg per cu m, do not interfere with the process but they can have a detrimental effect in a biological clarification plant or in a natural watercourse. According to the present invention the main flow is not changed with regard to its p~
value.
Corresponding to the redox potential measured in the measuring flow there is controlled, via an electronic regula-tor with P, PI or PID characteristics, a dosing pump which doses the corresponding amount of a commercial formaldehyde solution into the ring conduit of the main flow ahead of the gas washer. Diluted or gaseous formaldehyde can also be used instead of the commercial formaldehyde quantities. When using, for example, a pai~ of gold thallamide electrodes for this purpose, the redox potential in the measuring flow 1;~5856S
reaches a value ranging from approximately +400 mv + 50 mv to ~100 mv + 50 mV in a pH value range from 7 to 10, preferably 8 to 8.5 as soon as the reaction of the cyanide with the formaldehyde is completed.
A preferred rated redox value lies between +650 mv +
50 mV and +850 mV + 50 mv, particularly preferred at +700 mV +
50 mv at a pH value of 8 to 8.5. When using a silver elec-trode instead of a gold electrode the rated redox values are lower by approximately 100 mV on reacting the end product of the reaction between cyanogen and formaldehyde. The optimal ad~ustment of the rated value must be determined by a preced-ing small-scale test this also applies correspondingly to other combinations of electrode pairs. By adding the formaldehyde solution ahead of the gas washer and by measuring the redox potential in the measuring flow after the gas washes the hydrogen cyanide contained in the gas to be washed is immediately converted into glycolic acid nitrile on the one hand but a reaction of the hydrogen cyanide with metals while forming heavy metal cyanides which are difficult to detoxicate is avoided on the other. Furthermore, at a high content of ammonia in the wash water hexamethylene tetramine can be formed with formaldehyde. The hexamethylene tetramine again releases formaldehyde to the extent that it is required for the reaction with the hydrogen cyanide while forming glycolic acid nitrile.
For this reaction a reaction time of maximally 2 minutes is required. This can always be maintained by means of the arrangement of the measuring and dosing points in the process according to the present invention. Furthermore, in this manner it is assured that no free formaldehyde is present in the washing cycle since only the amount stoichiometric to the cyanide content is dosed in by the redox-controlled dosage 1;~58565 amount on the one hand and in the presence of high NH3 con-tents an unintended slight excess is converted into biologi-cally readily degradeable hexamethylene tetramine on the other. For the decomposition of the glycol nitrile formed an oxidative treatment of the washing cycle or of the wash water J removed from the circulation is carried out.
Primarily hydrogen peroxide is suitable as oxidizing agent. Hydrogen-peroxide solutions in concentrations of 10 to 70% by weight are preferably used. A further important effect attained by the process according to the present invention is the reduction of the discharge of hydrogen cyanide from the evaporation coolers. The cooling cycles operated in foundries must be fitted with evaporation coolers to maintain an ade-quate cooling water temperature. In a laboratory test adjusted to plant conditions it could be determined that a flue gas with 15 p.p.m. of hydrogen cyanide results from a blast-furnace-gas wash water loaded with 10 mg of CN per litre while the flue gas from a blast-furnace-gas wash water treated with formaldehyde and having the same initial cyanide concentration contained only 2 p.p.m. of hydrogen cyanide.
secause of the reconcentration of the wash water and the deposits in the gas washers associated therewith it is very favourable to remove contlnuously a small amount of wash water from the circulation. Instead of the washing cycle this discharged water is ad~usted with lime or alkali liquor to pH
value of 8.5 to 12.5, preferably 10.5 and controlled poten-tiometrically, for example, adding H202, until the redox potential of +700 mV + 50 mV, as measured with a pair of gold-thallium amalgam/thallium chloride electrodes is attained.
The change in potential thus occurring can be used for signalizing the end of the detoxication reaction or in the case of continuous dosing of oxidizing agent for switching off l;~S~35~5 the dosing operation. As mentioned hereinbefore, a biological treatment, for example, in a correspondingly adapted clarifl-cation plant, can follow.
The process according to the present invention is applicable primarily to inorganic cyanides obtained in wash waters of smelting operations, garbage pyrolyses and plants processing or producing hydrocyanic acid or cyanide. Since the measuring flow is branched off at the point of the strongest cyanide ions concentration of the wash water the addition of the formaldehyde into the washing cycle prior to entering the gas washer is always so regulated that this addi-tion also is adequate for varying concentrations of cyanide ions. This is a decisive advantage of the process according to the present invention. Furthermore, the gas washers can be operated in pH ranges in which the carbon dioxide of the air is not absorbed.
Heretofore, the operations had to be carried out in the pH range >13 to assure a complete separation of the hydro-gen cyanide and of the carbon dioxide. This is required since the carbonic acid is a stronger acid than the hydrocyanic acid and in the lower pH range carbon dioxide is preferably absorbed by the wash water. Thus, according to the process of the present invention a distinctly lower load of the wash water with neutral salts is caused than was the case hereto-fore. The pH range of the wash water does not have to be changed for reasons of an undesired absorption of gases, as for example, the absorption of carbon dioxide, nor for reasons of effectively carrying out the process according to the present invention. The pH range can vary within wide limits but it will usually be below pH = 7. However, in individual cases, as in the treatmen~t of flue gases and in the production 1;~5856S
of ferromanganese, values of between 10 and 11 are also possible.
All of these effluents can be used for the process according to the present invention with the pH value unchanged. The measuring flow alone is adjusted to a pH value of preferably 8 to 8.5. It should also be emphasized that even large amounts of ammonium ions sulphide ions, as they are encountered in effluents from the iron and steel industry and garbage pyrolyses, do not interfere with the measurement in the measuring flow (for the determination of the influence of ammonia on the required amount of formaldehyde see the Exam-ples 3 and 4). The following two principles of carrying out the process serve for further illustrating the process accord-ing to the present invention: these two procedures (see Figure 1 and 2) can be carried out individually or in combina-tion - adapted to any given case. Thus, for example, the method according to Figure 1 can be used instead of the wash-ing cycle in Figure 2, i.e., instead of the system '~gas washer - measuring flow - actual washing cycle". In this latter case dosing of oxidizing agent into the washing cycle is then dis-pensed with. In Figure 1 the washing of gas with formaldehyde and simultaneous oxidation with hydrogen peroxide in the cir-culation system is shown. The crude gas is fed via llne 1 to the gas washer 2 from below and the pure gas is removed via line 3. The wash water is fed via line 4 to the gas washer 2 from above in a counterflow and drawn off via line 16.
The suspended substances contained in the wash water are separated in the sedimentation tank 5, for example, a cir-cular reamer, and the wash water is then passed via the line 17 to the cooler 6, for example, an evaporation cooler, from where it gets back into the gas washer 2 via line 4.
125~356~
The washing cycle comprises the lines 4,16,17 as well as the gas washer 2, the sedimenta-tion tank 5 and the cooler 6. After the gas washer 2 a small measuring flow 7 is branched off from the washing cycle and controlled via a regu-lator 8. The pH value of the measuring flow is adjusted to, e.g., 8 to 8.5 by adding acid via line 9 or liquor via line 10. Corresponding to the redox potential also measured with the regulator 8 and to the deviation thus determined formalde-hyde is dosed via line 11 into the washing cycle prior to entering the washer 2. The silver compound, for example, nitrate, enters the measuring flow 7 via line 9a.
The redox potential of the washing cycle, whose pH
value is unchanged, is measured with the regulator 12 and cor-responding to the deviation determined hydrogen peroxide is dosed via line 13 after the circular reamer 5. The residence time in the cooler and in the entire washing cycle is utilized for the perhydrolysis of the glyconitrile formed in the gas washer 2. The water washed from the cooler 6, whose pH must be changed when required and is drawn off via line 14, is replaced by adding the corresponding amount of fresh water via line 15 after the cooler 6.
In Figure 2 the washing of the gas with formaldehyde and separate treatment of gas wash water with hydrogen per-oxide is shown. The crude gas is fed via line 1 to the gas washer 2 from below and the pure gas is drawn off via line 3.
The wash water is passed in a counterflow via line 4 to the gas washer 2 from above and drawn off via line 4a. The sus-pended substances contained in the wash water are removed in the sedimentation tank 5 and then passed to the detoxication plant.
After the gas washer 2 a small measuring flow 7 is taken out of the washing cycle via line 16 and controlled via ~58565 a regulator 8 by adding liquor via line 9 or acid via line 10 and adjusted to a pH value of 8 to 8.5 while adding a silver compound via line 9.
Corresponding to both the redox potential also measured with the regulator 8 and the deviation thus deter-mined formaldehyde is dosed via line 11 into the washing cycle, i.e., into line 4, prior to entering the washer 2. At the same time as much fresh water is added via pipe 12 ahead of the gas washer 2 as is washed away via line 5. The ratio of water washed away to the amount of water in the washing cycle is determined by both the degree of pollution and the degree of hardness. This can be d~termined by a preliminary test.
The water discharged from the sedimentation device 5 is first passed to the first reaction tank 13, which is pro-vided with a stirrer 14, a pH-measuring and control device 15 as well as with a liquor dosing device 16 and an H2O2 dosing device 17. When required, the pH of the effluent to be treated is first adjusted to a pH value >10.5. In the second reaction tank 18 provided with a stirrer 19 and a redox-measuring and control device 20 the corresponding amount of H2O2 is dosed via 17 into the first reaction tank 13 according to both the measured redox potential and the deviation thus determined. The size of the reaction tanks is such, that an adequate residence time corresponding to the effluent flow is attained in order to assure a complete perhydrolysis of the glyconitrile. The reaction is potentiometrically monitored via the regulator 20.
The oxidatively treated water is then passed into the third reaction tank 21, which is provided with a strirrer 22 a pH-measuring and control device 23 and an acid dosing l~S~5~jS
device 24. The effluent is then adjusted to a pH value pre-scribed for the discharge.
The present invention will be further illustrated by the following Examples:
The effect which can be attained by a potentiometri-cally controlled addition of formaldehyde is described in Example 1.
A washing cycle having a volume of 5 cu m and a cir-culation of 60 cu m per hour had the following composition:
directly argentometrically determinable cyanide83 mg of CN /litre total cyanide DIN 38 405 D13.1135 mg of CN /litre free N~3 3800 mg of NH3/litre total NH3 3800 mg of NH3/litre iron, total 46.5 mg of Fe/litre
2 cu m of water were added to and removed from the cycle per hour. By electrometrically controlled dosing of an average of 4B kg of (37% by weight) of H2CO per hour ahead of the gas washer the following analytical values were determined in the effluent after an operating time of 8 hours:
20 directly argentometrically determinable cyanide629 mg of CN /litre free NH3 490 mg of NH3/litre total NH3 3900 mg of NH3/litre iron, complex bonded 1.2 mg of Fe/litre iron, total 1.2 mg of Fe/litre It was the aim of this test to convert the hydrogen cyanide completely into glycolic acid nitrile and to reduce the free ammonia to <600 mg of NH3 per litre. The residual NH3 was to be converted into hexamethylene tetramine by the addition of H2C0. According to the analytical data an average of 52.7 kg of a 37% by we~ght H2CO would have had to be applied. The content of total cyanide according to DIN
38 405, D 13.1 could not be determined since the cyanide pre 1~8~65 sent as glycolic acid nitrile is only incompletely determined with this analytical method. From the content of complex bonded iron a content of 3.35 mg of CN- per litre is deter-mined as complex iron cyanide. A reduction of the initial content of 52 mg of CN to 3.35 mg of CN , i.e., by 95% has thus been attained. In other words the renewed formation of complex heavy metal cyanides has been prevented by the addi-tion of formaldehyde.
After the treatment of this effluent with H202 the content of total cyanide could be reduced to <1 mg of CN
/litre and the content of easily releasable cyanide according to DIN 38405, D13.2 to <0.1 mg of CN /litre.
In Example 2 the importance of the adjustment of the pH value for the exact formaldehyde dosage is to be shown as it will be carried out below in the Examples 5 and 6 on the measuring flow.
Example 2 was carried out without a measuring flow.
A washing cycle having a volume 4000 cu m and a cir-culation of 1400 cu m per hour had the following composition:
directly argentometrically determinable cyanide260 mg of CN /litre total cyanide DIN 38 405, D 13.1 325 mg of CN /litre easily releasable cyanide DIN 38 405, D 13.2 26.4 mg of CN /litre manganese 1. 2 mg of Mn/litre complex iron cyanide 110 mg of CN /litre Within 3 hours 2500 litres of 37% by weight of H2CO were added to the cycle, whereupon 20 kg of 37% by weight H2CO were con-tinuously dosed in.
As a function of the time the effluent, from which 20 cu m were removed per hour and replaced by fresh water, 1.;~58565 changed within 72 hours as follows:
TABLE
Time Cyanïde Cyanide Manganese Cyanide Addltion Hours Total mg Easily mg of Mn as Fe(CN)6 of ~ormal-Of CN- Releasable per litre mg of CN dehyde per litre mg of CN per litre _ ____ _________ per litre ___________ _____________ _________ .. 0.... ... 325... ... 264...... ... 1.2..... ... 11.0*..... .. 2750.kg .24.... ... 2~5... ... 187...... ... 0.7~ not deter.** ....... ..20.kg/hr .29.~ not deter not deter... ... -....... ... 77.5*..... .... 20.kg/hr .48.... ... 193... 1.. 156...... ... 0.4..... ... 42.6*..... .... 20.kg/hr ..... ......... I.. ........... ............. .. +950.kg .51.... ... 214... ... 171...... ... 0.3..... ... 40.6*..... .... 20.kg .72.... ... 272... ... 208...... ... 0.2..... ... 38.5*..... .... 20.kg ______ _________ __________ __________ _____________ _________ * = computed from the Fe content ~AAS = atom absorption spectroscopy) ** = not determined Within 48 hours a total of 4000 cu m of wash water +900 cu m of discharged water were treated with H2CO. At an average content of 250 mg of CN /litre the use of 3820 kg of 37% by weight H2CO would have been required. A total of 4850 kg were used. The slightly higher dosage amount is due to the fact the washing cycle had a pH value of 10.1 to 10.5 and was not changed. This resulted in an unfavourable effect on the redox potential and thus in the overdosage of H2CO relative to the cyanide content.
Although only approximately 22% of the cycle was replaced by fresh water, not only did the H2CO dosage result in a decisive improvement of the settling characteristic of the suspended substances but it also resulted in a reduction of the manganese content by more than 80% and of the content of complex iron cyanide by 65%.
The following Examples 3 and 4 are laboratory tests for determining the effect of high ammonia concentrations on the required dosage amount of formaldehyde.
4 litres of a solution with 300 mg of CN /litre, 4000 mg of NH3/litre and 1 g of Ag/litre were adjusted with 12585~5 25% by weight hydrochloric acid to a pH value of ~.5 and mixed in portions with 37% by weight H2CO. The redox potential measured with an Au/thallium amalgam-thallium chloride measur-ing chain increased even after adding 4% of the theoretical amount of H2CO, relative to the content of cyanide and ammo-nia, from +380 mV to +675 mV and remained practically unchanged until an addition of 30% of the theoretical amount of ~2CO. 4% of the theoretical amount of H2CO correspond to 126% of the theoretical amount, relative to cyanide; however, 30% of the theoretical amount of H2CO already correspond to 950% of the theoretical amount, relative to the cyanide.
From Example 3 it is evident that even in the case of high ammonia contents the formaldehyde applied is prefer-ably used for the conversion of cyanide. This is evident from the potential which no longer changes. The excess formalde-hyde is reacted to hexamethylene tetramine.
A solution according to Example 3 (pH value = 8 to 8.5) was mixed with 1.38 g of H2CO corresponding to 100% of the theoretical amount, relative to cyanide, and the course of the redox potential was measured. The initial potential of +380 mv at first increased rapidly to +580 mV; after a total of 2 minutes the final potential of +670 mV was reached. With only 100% of the theoretical amount the same final value is thus reached as in Example 3 with 950% of the theoretical amount. It thus is a case of an absolutely reliable control method.
By subsequently adding 120% of the theoretical amount of aqueous 50% by weight aqueous hydrogen peroxide, relative to the cyanide, the potential no longer changed.
Only by adding 3.8 times the amount of H2O2, relative to the lZ58~iS
cyanide, did the potential increase to +700 mv at pH 8.5 and remained constant for more than 2 hours.
It has thus been assured that the process for con-trolling the formaldehyde dosage also operates satisfactorily when large excesses of oxidizing agents are present in the pH
range of 8 to 8.5.
The solution thus treated was then adjusted with an aqueous 10% by weight solution of caustic soda to pH 10.5.
The potential then dropped to +400 mv. However, by perhydro-lysis of the glyconitrile with the H2O2 the potential increased to +750 mV within 10 minutes, indicating the end of the detoxication. With the pyridine-barbituric acid reagent it could be proved analytically that the cyanide content had been degraded to <0.1 mg of CN /litre.
Accordins to this Example it is required merely to branch off a small partial flow of the washing cycle and to condition it. The measurement of the redox potential for the automatic H2CO dosage at pH 8.5 can also be carried out in the presence of oxidizing agents, as for example, H2O2. When the pH automatically adjusts to a value of lO to ll, as for exam-ple, in the production of ferromanganese it thus is possible to carry out the oxidati~e treatment already in the cycle and, when required, to sub~ect the discharged water to an after treatment.
The Examples 5 and 6 are practical examples of the process according to the present invention corresponding to Figure 1 and 2.
A cooling cycle (Fig. 1) having a volume of 4200 cu m, a circulation of 800 cu m per hour and lO0 cu m of elutrition per hour had t~e following composition:
12 585~, total cyanide DIN 3a 405, D 13.1 78.5 mg of CN /litre easily releasable cyanide, DIN 38 405, D 13.2 67.4 mg of CN /litre ammonia 35 mg of NH3/litre pH value 9. 8 A measuring flow of 100 litre per hour was taken from this cooling cycle after the gas washer and continuously controlled via an electronic pH regulator and adjusted to pH 8 to 8. 5 by adding hydrochloric acid. At the same time a silver nitrate solution was added to maintain an Ag concentration of 1 ~ g per litre. By measuring the redox potential with a pair of Au/thallium amide-thallium chloride electrodes ahead of the gas washer - controlled by an electronic redox regulator - a 37~ by weight aqueous formaldehyde was added until a constant redox potential of +575 mv was maintained after the gas washer.
At the same time the redox potential was measured in the main flow after the gas washer with a second Au/thallium amide-thallium chloride measuring chain. Corresponding to the measured redox potential in the main flow ahead of the gas cooler at pH 9.8 H2O2 was added until a constant redox poten-tial of +750 mV was attained.
The following dosage amounts and effective compo-nents were determined formaldehyde dosage over 16 hours:
on the average 20.2 litres/hour (37% by weight) = 7 .12 kg of CN /h 100 cu m of effluent/hour with 67.4 mg of CN /litre = 6.74 kg of CN /h formaldehyde applied 105% of the theoretical amount, relative to CN
H2O dosage over 16 hours:
o~ the average 30 litres/hour ( 50% by weight) The cyanide content at the overflow after the gas cooler varied between 0.1 and 0.3 mg of CN- per litre. At the inlet at the plant outlet (residence time) values <0.1 mg of CN
/litre were always recorded.
12585~i~
A washing cycle (Fig. 2) having a volume of 5 cu m and a circulation of 60 cu m per hour had the following composition:
directly argentometrically determinable cyanide535 mg of CN /litre (average over 16 hours) ammonia 3650 mg of NH3/litre (average over 16 hours) pH value 7.9 A measuring flow of lO0 litre per hour was taken from the cycle ahead of the washer and continuously controlled via an electronic pH regulator and adjusted to pH 8 to 8.5 by adding a 10% by weight solution of caustic soda. At the same time a silver nitrate solution was added in order to maintain an Ag concentration of l ~ of Ag per litre. By measuring the redox potential (with a pair of Au/thallium amide-thallium chloride electrodes a 37% by weight H2CO, controlled by an electronic redox regulator, was added ahead of the gas washer until a constant redox potential of +675 mV was maintained after the gas washer. 2 cu m of wash water per hour were replaced by fresh water. Because of this arrangement of the measuring and dosing device the following dosage amount of 37% by weight H2CO was required:
20 directly argentometrically determinable cyanide629 mg of CN /litre free NH3 490 mg of NH3/litre total NH3 3900 mg of NH3/litre iron, complex bonded 1.2 mg of Fe/litre iron, total 1.2 mg of Fe/litre It was the aim of this test to convert the hydrogen cyanide completely into glycolic acid nitrile and to reduce the free ammonia to <600 mg of NH3 per litre. The residual NH3 was to be converted into hexamethylene tetramine by the addition of H2C0. According to the analytical data an average of 52.7 kg of a 37% by we~ght H2CO would have had to be applied. The content of total cyanide according to DIN
38 405, D 13.1 could not be determined since the cyanide pre 1~8~65 sent as glycolic acid nitrile is only incompletely determined with this analytical method. From the content of complex bonded iron a content of 3.35 mg of CN- per litre is deter-mined as complex iron cyanide. A reduction of the initial content of 52 mg of CN to 3.35 mg of CN , i.e., by 95% has thus been attained. In other words the renewed formation of complex heavy metal cyanides has been prevented by the addi-tion of formaldehyde.
After the treatment of this effluent with H202 the content of total cyanide could be reduced to <1 mg of CN
/litre and the content of easily releasable cyanide according to DIN 38405, D13.2 to <0.1 mg of CN /litre.
In Example 2 the importance of the adjustment of the pH value for the exact formaldehyde dosage is to be shown as it will be carried out below in the Examples 5 and 6 on the measuring flow.
Example 2 was carried out without a measuring flow.
A washing cycle having a volume 4000 cu m and a cir-culation of 1400 cu m per hour had the following composition:
directly argentometrically determinable cyanide260 mg of CN /litre total cyanide DIN 38 405, D 13.1 325 mg of CN /litre easily releasable cyanide DIN 38 405, D 13.2 26.4 mg of CN /litre manganese 1. 2 mg of Mn/litre complex iron cyanide 110 mg of CN /litre Within 3 hours 2500 litres of 37% by weight of H2CO were added to the cycle, whereupon 20 kg of 37% by weight H2CO were con-tinuously dosed in.
As a function of the time the effluent, from which 20 cu m were removed per hour and replaced by fresh water, 1.;~58565 changed within 72 hours as follows:
TABLE
Time Cyanïde Cyanide Manganese Cyanide Addltion Hours Total mg Easily mg of Mn as Fe(CN)6 of ~ormal-Of CN- Releasable per litre mg of CN dehyde per litre mg of CN per litre _ ____ _________ per litre ___________ _____________ _________ .. 0.... ... 325... ... 264...... ... 1.2..... ... 11.0*..... .. 2750.kg .24.... ... 2~5... ... 187...... ... 0.7~ not deter.** ....... ..20.kg/hr .29.~ not deter not deter... ... -....... ... 77.5*..... .... 20.kg/hr .48.... ... 193... 1.. 156...... ... 0.4..... ... 42.6*..... .... 20.kg/hr ..... ......... I.. ........... ............. .. +950.kg .51.... ... 214... ... 171...... ... 0.3..... ... 40.6*..... .... 20.kg .72.... ... 272... ... 208...... ... 0.2..... ... 38.5*..... .... 20.kg ______ _________ __________ __________ _____________ _________ * = computed from the Fe content ~AAS = atom absorption spectroscopy) ** = not determined Within 48 hours a total of 4000 cu m of wash water +900 cu m of discharged water were treated with H2CO. At an average content of 250 mg of CN /litre the use of 3820 kg of 37% by weight H2CO would have been required. A total of 4850 kg were used. The slightly higher dosage amount is due to the fact the washing cycle had a pH value of 10.1 to 10.5 and was not changed. This resulted in an unfavourable effect on the redox potential and thus in the overdosage of H2CO relative to the cyanide content.
Although only approximately 22% of the cycle was replaced by fresh water, not only did the H2CO dosage result in a decisive improvement of the settling characteristic of the suspended substances but it also resulted in a reduction of the manganese content by more than 80% and of the content of complex iron cyanide by 65%.
The following Examples 3 and 4 are laboratory tests for determining the effect of high ammonia concentrations on the required dosage amount of formaldehyde.
4 litres of a solution with 300 mg of CN /litre, 4000 mg of NH3/litre and 1 g of Ag/litre were adjusted with 12585~5 25% by weight hydrochloric acid to a pH value of ~.5 and mixed in portions with 37% by weight H2CO. The redox potential measured with an Au/thallium amalgam-thallium chloride measur-ing chain increased even after adding 4% of the theoretical amount of H2CO, relative to the content of cyanide and ammo-nia, from +380 mV to +675 mV and remained practically unchanged until an addition of 30% of the theoretical amount of ~2CO. 4% of the theoretical amount of H2CO correspond to 126% of the theoretical amount, relative to cyanide; however, 30% of the theoretical amount of H2CO already correspond to 950% of the theoretical amount, relative to the cyanide.
From Example 3 it is evident that even in the case of high ammonia contents the formaldehyde applied is prefer-ably used for the conversion of cyanide. This is evident from the potential which no longer changes. The excess formalde-hyde is reacted to hexamethylene tetramine.
A solution according to Example 3 (pH value = 8 to 8.5) was mixed with 1.38 g of H2CO corresponding to 100% of the theoretical amount, relative to cyanide, and the course of the redox potential was measured. The initial potential of +380 mv at first increased rapidly to +580 mV; after a total of 2 minutes the final potential of +670 mV was reached. With only 100% of the theoretical amount the same final value is thus reached as in Example 3 with 950% of the theoretical amount. It thus is a case of an absolutely reliable control method.
By subsequently adding 120% of the theoretical amount of aqueous 50% by weight aqueous hydrogen peroxide, relative to the cyanide, the potential no longer changed.
Only by adding 3.8 times the amount of H2O2, relative to the lZ58~iS
cyanide, did the potential increase to +700 mv at pH 8.5 and remained constant for more than 2 hours.
It has thus been assured that the process for con-trolling the formaldehyde dosage also operates satisfactorily when large excesses of oxidizing agents are present in the pH
range of 8 to 8.5.
The solution thus treated was then adjusted with an aqueous 10% by weight solution of caustic soda to pH 10.5.
The potential then dropped to +400 mv. However, by perhydro-lysis of the glyconitrile with the H2O2 the potential increased to +750 mV within 10 minutes, indicating the end of the detoxication. With the pyridine-barbituric acid reagent it could be proved analytically that the cyanide content had been degraded to <0.1 mg of CN /litre.
Accordins to this Example it is required merely to branch off a small partial flow of the washing cycle and to condition it. The measurement of the redox potential for the automatic H2CO dosage at pH 8.5 can also be carried out in the presence of oxidizing agents, as for example, H2O2. When the pH automatically adjusts to a value of lO to ll, as for exam-ple, in the production of ferromanganese it thus is possible to carry out the oxidati~e treatment already in the cycle and, when required, to sub~ect the discharged water to an after treatment.
The Examples 5 and 6 are practical examples of the process according to the present invention corresponding to Figure 1 and 2.
A cooling cycle (Fig. 1) having a volume of 4200 cu m, a circulation of 800 cu m per hour and lO0 cu m of elutrition per hour had t~e following composition:
12 585~, total cyanide DIN 3a 405, D 13.1 78.5 mg of CN /litre easily releasable cyanide, DIN 38 405, D 13.2 67.4 mg of CN /litre ammonia 35 mg of NH3/litre pH value 9. 8 A measuring flow of 100 litre per hour was taken from this cooling cycle after the gas washer and continuously controlled via an electronic pH regulator and adjusted to pH 8 to 8. 5 by adding hydrochloric acid. At the same time a silver nitrate solution was added to maintain an Ag concentration of 1 ~ g per litre. By measuring the redox potential with a pair of Au/thallium amide-thallium chloride electrodes ahead of the gas washer - controlled by an electronic redox regulator - a 37~ by weight aqueous formaldehyde was added until a constant redox potential of +575 mv was maintained after the gas washer.
At the same time the redox potential was measured in the main flow after the gas washer with a second Au/thallium amide-thallium chloride measuring chain. Corresponding to the measured redox potential in the main flow ahead of the gas cooler at pH 9.8 H2O2 was added until a constant redox poten-tial of +750 mV was attained.
The following dosage amounts and effective compo-nents were determined formaldehyde dosage over 16 hours:
on the average 20.2 litres/hour (37% by weight) = 7 .12 kg of CN /h 100 cu m of effluent/hour with 67.4 mg of CN /litre = 6.74 kg of CN /h formaldehyde applied 105% of the theoretical amount, relative to CN
H2O dosage over 16 hours:
o~ the average 30 litres/hour ( 50% by weight) The cyanide content at the overflow after the gas cooler varied between 0.1 and 0.3 mg of CN- per litre. At the inlet at the plant outlet (residence time) values <0.1 mg of CN
/litre were always recorded.
12585~i~
A washing cycle (Fig. 2) having a volume of 5 cu m and a circulation of 60 cu m per hour had the following composition:
directly argentometrically determinable cyanide535 mg of CN /litre (average over 16 hours) ammonia 3650 mg of NH3/litre (average over 16 hours) pH value 7.9 A measuring flow of lO0 litre per hour was taken from the cycle ahead of the washer and continuously controlled via an electronic pH regulator and adjusted to pH 8 to 8.5 by adding a 10% by weight solution of caustic soda. At the same time a silver nitrate solution was added in order to maintain an Ag concentration of l ~ of Ag per litre. By measuring the redox potential (with a pair of Au/thallium amide-thallium chloride electrodes a 37% by weight H2CO, controlled by an electronic redox regulator, was added ahead of the gas washer until a constant redox potential of +675 mV was maintained after the gas washer. 2 cu m of wash water per hour were replaced by fresh water. Because of this arrangement of the measuring and dosing device the following dosage amount of 37% by weight H2CO was required:
3.2 litres of 37% by weight H2CO were added per hour on the average, i.e., 106% of theoretical amount, relative to the cyanide content of the discharged wash water.
After separating the suspended substances the dis-charged wash water was first adjusted with a solution of caus-tic soda to pH value of 10.5, whereupon by redox measurement with a pair of Au/thallium amide-thallium chlo~ide electrodes - controlled via an electronic regulator - a 50% by weight H2O2 was added until the redox potential had risen to +750 mV
and was held at this value during the continuous effluent 1;:585~iS
treatment. After a total reaction time of 3.5 hours it was possible to reneutralize to a pH value of 8.5 corresponding to the ~iven plant size. The cyanide content in the first reac-tion tank (reaction time approximately 1 hour) always was <0.1 mg of CN /litre.
After separating the suspended substances the dis-charged wash water was first adjusted with a solution of caus-tic soda to pH value of 10.5, whereupon by redox measurement with a pair of Au/thallium amide-thallium chlo~ide electrodes - controlled via an electronic regulator - a 50% by weight H2O2 was added until the redox potential had risen to +750 mV
and was held at this value during the continuous effluent 1;:585~iS
treatment. After a total reaction time of 3.5 hours it was possible to reneutralize to a pH value of 8.5 corresponding to the ~iven plant size. The cyanide content in the first reac-tion tank (reaction time approximately 1 hour) always was <0.1 mg of CN /litre.
Claims (22)
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. In a process for the purification of industrial gas or waste gases which contain hydrogen cyanide by a circulating wash with water and formaldehyde and a simultaneous oxidative treatment or a subsequent oxidative treatment of the circulating water removed from the circuit the improvement comprising dosing the formaldehyde into the gas wash circuit before the gas washer and controlling the amount of formaldehyde added so that it is stoichiometrically equal to the cyanide content by continuously measuring the redox potential with a pair of electrode consisting essentially of a noble metal and a reference electrode in a measuring side stream branched off after the washer and controlling the pH
of the side stream to a pH between 7 and 1 by dosing in alkali or acid in the presence of at least 1 ppb of silver ions.
of the side stream to a pH between 7 and 1 by dosing in alkali or acid in the presence of at least 1 ppb of silver ions.
2. A process according to claim 1 wherein the industrial or waste gas employed contains ammonia.
3. A process according to claim 1 wherein the industrial or waste gas employed contains at least one member of the group consisting of carbon oxides, hydrocarbons, hydrogen sulfide, sulfur oxides, and heavy metal containing dust.
4. A process according to claim 1 wherein the pH is controlled to 8.0 to 8.5.
5. A process according to claim 1 wherein there is employed a gold-thalamide electrode pair and the redox nominal value at a pH of 7 to 10 is adjusted to +400 mV?50 mV to +1000 mV?50 mV.
6. A process according to claim 1 wherein there is employed a gold-thalamide electrode pair and the redox nominal value at a pH of 8 to 8.5 is adjusted to +400 mV?50 mV to +1000 mV?50 mV.
7. A process according to claim 6 wherein the redox nominal value is adjusted to +650 mV?50 mV to +850 mV?50 mV.
8. A process according to claim 7 wherein the redox nominal value is adjusted to 700 mV?50 mV.
9. A process according to claim 8 wherein the measuring stream contains 1 ppb of silver ions.
lo. A process according to claim 7 wherein the measuring stream contains 1 ppb of silver ions.
11. A process according to claim 7 wherein the measuring stream contains a maximum of 10 ppb of silver ions.
12. A process according to claim 1 wherein the measuring stream contains a maximum of 10 ppb of silver ions.
13. A process according to claim 12 comprising removing a portion of the wash water from the circuit after the formaldehyde treatment, establishing a pH of 8.5 to 12.5 in the removed wash water and treating the removed wash water with sufficient oxidizing agent to completely hydrolyze the glycolonitrile formed in the formaldehyde treatment.
14. A process according to claim 13 wherein the oxidizing agent is hydrogen peroxide.
15. A process according to claim 9 comprising removing a portion of the wash water from the circuit after the formaldehyde treatment, establishing a pH of 8.5 to 12.5 in the removed wash water and treating the removed wash water with sufficient oxidizing agent to completely hydrolyze the glycolonitrile formed in the formaldehyde treatment.
16. A process according to claim 15 wherein the oxidizing agent is hydrogen peroxide.
17. A process according to claim 7 comprising removing a portion of the wash water from the circuit after the formaldehyde treatment, establishing a pH of 8.5 to 12.5 in the removed wash water and treating the removed wash water with sufficient oxidizing agent to completely hydrolyze the glycolonitrile formed in the formaldehyde treatment.
18. A process according to claim 17 wherein the oxidizing agent is hydrogen peroxide.
19. A process according to claim 1 comprising removing a portion of the wash water from the circuit after the formaldehyde treatment, establishing a pH of 8.5 to 12.5 in the removed wash water and treating the removed wash water with sufficient oxidizing agent to completely hydrolyze the glycolonitrile formed in the formaldehyde treatment.
20. A process according to claim 19 wherein the oxidizing agent is hydrogen peroxide.
21. A process according to claim 19 including the steps of continuously measuring the redox potential in the main stream after the gas washer and dosing into the main stream before the gas cooler sufficient oxidizing agent to maintain the redox potential constant.
22. A process according to claim 21 wherein the oxidizing agent is hydrogen peroxide.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DEP3534677.9 | 1985-09-28 | ||
| DE3534677A DE3534677C1 (en) | 1985-09-28 | 1985-09-28 | Process for cleaning industrial gases |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1258565A true CA1258565A (en) | 1989-08-22 |
Family
ID=6282224
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA000519215A Expired CA1258565A (en) | 1985-09-28 | 1986-09-26 | Process for purifying industrial gases and industrial flue gases |
Country Status (7)
| Country | Link |
|---|---|
| US (1) | US4731232A (en) |
| EP (1) | EP0223904B1 (en) |
| JP (1) | JPS6274438A (en) |
| AT (1) | ATE51158T1 (en) |
| CA (1) | CA1258565A (en) |
| DE (2) | DE3534677C1 (en) |
| ES (1) | ES8703747A1 (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE3820899C1 (en) * | 1988-06-21 | 1989-11-09 | Degussa Ag, 6000 Frankfurt, De | Method and apparatus for the continuous measurement and adjustment of the NH3 leakage in the waste stream from waste gas denitrogenation by means of ammonia |
| DE4110056A1 (en) * | 1991-03-27 | 1992-10-01 | Degussa | METHOD FOR DETOXIFYING AQUEOUS SOLUTIONS CONTAINING CYANHYDRINE AND / OR NITRILE |
| DE4116840A1 (en) * | 1991-05-23 | 1992-11-26 | Degussa | METHOD FOR SEPARATING HYDROGEN CYANIDE FROM GASES AND EXHAUST GASES |
| US5217623A (en) * | 1991-05-30 | 1993-06-08 | Northwestern Flavors, Inc. | Method of detoxifying cyanide waste |
| DE4344598A1 (en) * | 1993-12-24 | 1995-06-29 | Degussa | Method for controlling the detoxification of cyanide effluents |
| FR2740355B1 (en) * | 1995-10-27 | 1998-01-16 | Cos Ingenierie | PROCESS FOR AMMONIACAL DEPOLLUTION OF LIQUID OR GASEOUS EFFLUENTS |
| US5810943A (en) * | 1996-11-27 | 1998-09-22 | Caron; Daniel | Washer apparatus with waste water pH neutralization system and method for the pH neutralization of waste water |
| AT508502B1 (en) * | 2009-08-03 | 2011-03-15 | Voestalpine Stahl Gmbh | METHOD AND DEVICE FOR CONTINUOUS DETERMINATION OF THE LEVEL OF AT LEAST ONE CN COMPOUND IN AN AQUEOUS SOLUTION |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US2859090A (en) * | 1955-02-07 | 1958-11-04 | Exxon Research Engineering Co | Treatment of fluids for removal of hydrogen cyanide |
| DE1069403B (en) * | 1957-04-26 | 1959-11-19 | ||
| DE1523018A1 (en) * | 1964-10-08 | 1969-04-24 | Siemens Ag | Measuring device for the automatic measurement of the cyanide content and / or the pH value of industrial waste water |
| US3923608A (en) * | 1972-08-10 | 1975-12-02 | Orion Research | Method of detecting low levels of cyanide |
| DE2254682A1 (en) * | 1972-11-08 | 1974-05-22 | Nihon Filter Co Ltd | Treating effluents contg. chromates or cyanides - using redox potentiometer with pH correction circuit to avoid inadequate or excess dosing with chemicals |
| FR2267984B1 (en) * | 1974-04-22 | 1976-12-17 | Degussa | |
| DE2741929C2 (en) * | 1977-09-17 | 1982-08-12 | UOP Inc. GmbH, 5272 Wipperfürth | Continuous process for removing formaldehyde from industrial exhaust gases |
| US4211646A (en) * | 1977-10-31 | 1980-07-08 | Texaco Inc. | Waste water process |
| DE3412581A1 (en) * | 1984-04-04 | 1985-10-24 | KPA Kiener Pyrolyse Gesellschaft für thermische Abfallverwertung mbH, 7000 Stuttgart | METHOD FOR PURIFYING GAS FROM PYROLYSIS PLANTS OF WASTE |
-
1985
- 1985-09-28 DE DE3534677A patent/DE3534677C1/en not_active Expired
-
1986
- 1986-05-13 EP EP86106469A patent/EP0223904B1/en not_active Expired - Lifetime
- 1986-05-13 AT AT86106469T patent/ATE51158T1/en not_active IP Right Cessation
- 1986-05-13 DE DE8686106469T patent/DE3669662D1/en not_active Expired - Lifetime
- 1986-05-19 ES ES555087A patent/ES8703747A1/en not_active Expired
- 1986-07-23 US US06/888,568 patent/US4731232A/en not_active Expired - Lifetime
- 1986-09-26 CA CA000519215A patent/CA1258565A/en not_active Expired
- 1986-09-29 JP JP61228543A patent/JPS6274438A/en active Granted
Also Published As
| Publication number | Publication date |
|---|---|
| ATE51158T1 (en) | 1990-04-15 |
| ES555087A0 (en) | 1987-03-01 |
| DE3534677C1 (en) | 1987-04-30 |
| EP0223904A3 (en) | 1988-04-13 |
| EP0223904A2 (en) | 1987-06-03 |
| DE3669662D1 (en) | 1990-04-26 |
| JPH0578368B2 (en) | 1993-10-28 |
| ES8703747A1 (en) | 1987-03-01 |
| EP0223904B1 (en) | 1990-03-21 |
| JPS6274438A (en) | 1987-04-06 |
| US4731232A (en) | 1988-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA1176387A (en) | Process for treating continuous effluent streams having varying contents of different oxidizable substances with hydrogen peroxide | |
| US6942840B1 (en) | Method for removal and stabilization of mercury in mercury-containing gas streams | |
| US4569769A (en) | Wastewater treatment | |
| JP7290511B2 (en) | Cyanogen-containing water treatment method and treatment equipment | |
| CA1258565A (en) | Process for purifying industrial gases and industrial flue gases | |
| SK280506B6 (en) | Anaerobic removal of sulphur compounds from wastewater | |
| JP4291151B2 (en) | Cyan waste liquid treatment method | |
| US4915849A (en) | Process for the treatment of effluents containing cyanide and other oxidizable substances | |
| US5667760A (en) | Methods for sweetening hydrocarbons | |
| EP0518585B1 (en) | Flue gas desulfurization process | |
| JP3572233B2 (en) | Flue gas desulfurization method and flue gas desulfurization system | |
| RU2743954C1 (en) | Waste water treatment method | |
| US3929636A (en) | Process for conditioning effluent contaminated by aldehyde compounds | |
| JP7157192B2 (en) | water treatment method | |
| Luzin et al. | Removal of cyanides from blast-furnace gas and wastewater | |
| US5200161A (en) | Process for the separation of hydrogen cyanide from gases and waste gases | |
| CN109516613A (en) | A kind of cyanide containing wastewater plating pretreating process | |
| JP7297512B2 (en) | Wastewater treatment method and wastewater treatment system | |
| WO1991000248A1 (en) | Process for reducing the cyanide content of a solution | |
| JP7454096B1 (en) | Wastewater treatment method | |
| Tan et al. | Chemical oxidation methods for the treatment of thiosalt-containing mill effluents | |
| RU2054307C1 (en) | Method of furnace gas scrubbing in sodium sulfide production from hydrogen sulfide and sulfur dioxide | |
| SU1368267A1 (en) | Method of removing dithionate-ions from waste water | |
| JPH05115887A (en) | Treatment for waste water containing inorganic sulfur-containing compound | |
| FI62040C (en) | PROCEDURE FOR THE FIELD OF THE METALLURGICAL INDUSTRY |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MKEX | Expiry |